Filing, Versions and the Misfiled Document

Part of Clinical Trial Assistant (CTA) Training: The Trial Master File.

About 4 minutes. Free, in your language, no account needed.

Why this matters

Most TMF findings are not missing documents. They are documents that are present, and wrong: the superseded version still active, the unsigned draft where the executed copy should be, the right document in the wrong study.

These are harder to find than absences, because a completeness report counts them as present.

Version control is the whole job

Every controlled document has a version and an effective date, and at any moment exactly one version is in force. Your file must make it obvious which.

Three habits prevent most version findings.

Supersede explicitly. When a new version arrives, the old one is marked superseded on the same day, not removed. Regulations expect the history; they do not accept ambiguity about what was current.

Check the effective date, not the arrival date. An amendment approved on 9 March and received on 3 April was in force from the approval, subject to your study's implementation rules. Filing it under April can misrepresent when the site should have been following it.

Match versions across the set. If protocol v4.0 is in force, the consent form in the file should be the version approved against v4.0. A mismatch between those two is one of the most consequential findings there is, because it can mean people consented to a study that no longer exists.

The misfiled document

A document in the wrong place has three failure modes, and they are not equally serious.

Filed in the wrong section of the right study: irritating, findable, usually a completeness report will flag the gap it left behind.

Filed in the wrong study: serious. Two files are now wrong, one missing the document and one containing a document that does not belong. If it contains participant information, you may also have a privacy issue.

Filed where it does not belong at all, typically a shared drive or an inbox: worst, because nothing will ever flag it. The document exists, the responsible person believes it was handled, and the file says otherwise.

When you find a misfile, fix the file and record what you did. Moving a document quietly leaves an audit trail that looks like tampering even when it was a correction.

What to do when you cannot tell which version was in force

Do not guess, and do not pick the most likely one. Escalate with the specific question: which version was in force on this date, and what is the evidence. Someone will have it, usually the ethics correspondence or the sponsor's approval record.

An unresolved version question that is documented is a manageable problem. A version question that someone resolved by assumption is a finding waiting to be discovered.

A worked example

A site has been enrolling since January. In March, protocol v3.0 is approved and a new consent form v3.0 goes with it. In June, you are reconciling the file and find consent form v2.0 still marked current in the site's folder, alongside v3.0.

Three questions, in order.

When did v3.0 take effect at this site? From their ethics approval date, which you find is 2 April.

Did anyone consent on v2.0 after 2 April? This is the question that matters, and the file answers it: you compare consent dates against versions used. If the answer is no, you have a filing problem. If the answer is yes, you have participants consented on a superseded form, which is a deviation with a participant consequence and needs reporting today.

Why did nobody notice for ten weeks? Usually because the new version was sent and the old one was never explicitly superseded, so both looked current.

The fix and the record

Mark v2.0 superseded with the date, keep it in the file, and make the current version unambiguous.

Then write down what you found and when, even if the answer to question two was no. A version confusion that was investigated and closed cleanly is evidence the file is being managed. The same confusion found by an inspector, with no record of anyone having looked, reads very differently.

Resist the urge to simply remove the old version to tidy up. The history is part of the evidence, and a document that disappears from a controlled file without explanation looks worse than one that is clearly marked superseded.

Key takeaways